[PATCH 3/4] trace-cmd: Add a new option in trace.dat file for the address to function name mapping

From: "Tzvetomir Stoyanov (VMware)" <tz.stoyanov@gmail.com>
To: rostedt@goodmis.org
Cc: linux-trace-devel@vger.kernel.org
Subject: [PATCH 3/4] trace-cmd: Add a new option in trace.dat file for the address to function name mapping
Date: Fri,  7 Aug 2020 15:06:54 +0300	[thread overview]
Message-ID: <20200807120655.797084-4-tz.stoyanov@gmail.com> (raw)
In-Reply-To: <20200807120655.797084-1-tz.stoyanov@gmail.com>

There are cases where resolving a user space address recorded in the
trace file to function name is required, when this file is visualised.
This mapping is available during the trace. A new trace.dat file option
TRACECMD_OPTION_PID_SYMBOLS is added, for storing the mapping in the
file. A new API is introduced, to resolve address to name, using
information stored in the TRACECMD_OPTION_PID_SYMBOLS option:

 struct tracecmd_debug_symbols *
 tracecmd_search_task_symbol(struct tracecmd_input *handle,
			     int pid, unsigned long long addr);

Signed-off-by: Tzvetomir Stoyanov (VMware) <tz.stoyanov@gmail.com>
---
 include/trace-cmd/trace-cmd.h |   5 ++
 lib/trace-cmd/trace-input.c   | 144 ++++++++++++++++++++++++++++++++++
 tracecmd/trace-dump.c         |   3 +
 3 files changed, 152 insertions(+)

diff --git a/include/trace-cmd/trace-cmd.h b/include/trace-cmd/trace-cmd.h
index 5ebd076e..f47f6a19 100644
--- a/include/trace-cmd/trace-cmd.h
+++ b/include/trace-cmd/trace-cmd.h
@@ -112,6 +112,7 @@ enum {
 	TRACECMD_OPTION_TRACEID,
 	TRACECMD_OPTION_TIME_SHIFT,
 	TRACECMD_OPTION_GUEST,
+	TRACECMD_OPTION_PID_SYMBOLS,
 };
 
 enum {
@@ -262,6 +263,10 @@ unsigned int tracecmd_record_ts_delta(struct tracecmd_input *handle,
 struct tracecmd_proc_addr_map *
 tracecmd_search_task_map(struct tracecmd_input *handle,
 			 int pid, unsigned long long addr);
+
+struct tracecmd_debug_symbols *
+tracecmd_search_task_symbol(struct tracecmd_input *handle,
+			    int pid, unsigned long long addr);
 #ifndef SWIG
 /* hack for function graph work around */
 extern __thread struct tracecmd_input *tracecmd_curr_thread_handle;
diff --git a/lib/trace-cmd/trace-input.c b/lib/trace-cmd/trace-input.c
index af97883e..39da0cb8 100644
--- a/lib/trace-cmd/trace-input.c
+++ b/lib/trace-cmd/trace-input.c
@@ -19,6 +19,8 @@
 #include "trace-local.h"
 #include "kbuffer.h"
 #include "list.h"
+#include "trace-hash.h"
+#include "trace-hash-local.h"
 
 #define _STRINGIFY(x) #x
 #define STRINGIFY(x) _STRINGIFY(x)
@@ -98,6 +100,18 @@ struct host_trace_info {
 	struct ts_offset_sample	*ts_samples;
 };
 
+struct pid_function_map {
+	struct trace_hash_item hash;
+	struct tracecmd_debug_symbols symb;
+};
+
+struct pid_symbol_maps {
+	struct pid_symbol_maps *next;
+	int pid;
+	int symb_count;
+	struct trace_hash symbols;
+};
+
 struct tracecmd_input {
 	struct tep_handle	*pevent;
 	struct tep_plugin_list	*plugin_list;
@@ -130,6 +144,7 @@ struct tracecmd_input {
 
 	struct hook_list	*hooks;
 	struct pid_addr_maps	*pid_maps;
+	struct pid_symbol_maps	*pid_symbols;
 	/* file information */
 	size_t			header_files_start;
 	size_t			ftrace_files_start;
@@ -2430,6 +2445,92 @@ static int trace_pid_map_search(const void *a, const void *b)
 	return 0;
 }
 
+static void trace_pid_symbols_free(struct pid_symbol_maps *maps)
+{
+	struct pid_symbol_maps *del;
+	struct pid_function_map *fmap;
+	struct trace_hash_item **bucket;
+	struct trace_hash_item *item;
+
+	while (maps) {
+		del = maps;
+		maps = maps->next;
+		trace_hash_for_each_bucket(bucket, &del->symbols) {
+			trace_hash_while_item(item, bucket) {
+				trace_hash_del(item);
+				fmap = (struct pid_function_map *)item;
+				free(fmap->symb.name);
+				free(fmap->symb.fname);
+				free(fmap);
+			}
+		}
+		trace_hash_free(&del->symbols);
+	}
+}
+
+#define MAX_FUNC_NAME	256
+static int trace_pid_symbols_load(struct tracecmd_input *handle, char *buf)
+{
+	struct pid_symbol_maps *maps = NULL;
+	struct pid_function_map *s;
+	char file[PATH_MAX];
+	char fname[MAX_FUNC_NAME];
+	char *line;
+	int res;
+	int ret;
+	int i;
+
+	maps = calloc(1, sizeof(*maps));
+	if (!maps)
+		return -ENOMEM;
+
+	ret  = -EINVAL;
+	line = strchr(buf, '\n');
+	if (!line)
+		goto out_fail;
+
+	*line = '\0';
+	res = sscanf(buf, "%x %d", &maps->pid, &maps->symb_count);
+	if (res != 2)
+		goto out_fail;
+
+	ret  = -ENOMEM;
+	if (trace_hash_init(&maps->symbols, 64) < 0)
+		goto out_fail;
+
+	buf = line + 1;
+	line = strchr(buf, '\n');
+	for (i = 0; i < maps->symb_count; i++) {
+		if (!line)
+			break;
+		*line = '\0';
+		s = calloc(1, sizeof(*s));
+		if (!s)
+			goto out_fail;
+		res = sscanf(buf, "%"STRINGIFY(PATH_MAX)"s %"STRINGIFY(MAX_FUNC_NAME)"s %llx %llx",
+				file, fname, &s->symb.vma_start, &s->symb.vma_near);
+		if (res != 4)
+			break;
+		s->symb.fname = strdup(file);
+		s->symb.name = strdup(fname);
+		if (!s->symb.fname || !s->symb.name)
+			goto out_fail;
+		s->hash.key = trace_hash(s->symb.vma_near);
+		trace_hash_add(&maps->symbols, (struct trace_hash_item *)s);
+		buf = line + 1;
+		line = strchr(buf, '\n');
+	}
+
+	maps->next = handle->pid_symbols;
+	handle->pid_symbols = maps;
+
+	return 0;
+
+out_fail:
+	trace_pid_symbols_free(maps);
+	return ret;
+}
+
 /**
  * tracecmd_search_task_map - Search task memory address map
  * @handle: input handle to the trace.dat file
@@ -2469,6 +2570,43 @@ tracecmd_search_task_map(struct tracecmd_input *handle,
 	return lib;
 }
 
+/**
+ * tracecmd_search_task_symbol - Resolve address to function name
+ * @handle: input handle to the trace.dat file
+ * @pid: pid of the task, which function information is stored in the trace.dat
+ * @addr: address of the function
+ *
+ * Mapping of some functions to addresses of traced PIDs can be saved in the
+ * trace.dat file, using the "perf" sub command. If there is such information,
+ * this API can be used to look up into this function maps to find the name of
+ * the function and the name of the file where that function is loaded.
+ *
+ * A pointer to struct tracecmd_debug_symbols is returned, containing information
+ * about the reuquested @addr: the name of the function, its start address, the
+ * name of the binary file that contains the function.
+ */
+struct tracecmd_debug_symbols *
+tracecmd_search_task_symbol(struct tracecmd_input *handle,
+			    int pid, unsigned long long addr)
+{
+	struct pid_function_map *symb = NULL;
+	struct pid_symbol_maps	*map;
+
+	for (map = handle->pid_symbols; map; map = map->next) {
+		if (pid == map->pid)
+			break;
+	}
+
+	if (map)
+		symb = (struct pid_function_map *)trace_hash_find(&map->symbols,
+								  trace_hash(addr), NULL, NULL);
+
+	if (symb)
+		return &symb->symb;
+
+	return NULL;
+}
+
 static int handle_options(struct tracecmd_input *handle)
 {
 	long long offset;
@@ -2610,6 +2748,10 @@ static int handle_options(struct tracecmd_input *handle)
 		case TRACECMD_OPTION_GUEST:
 			trace_guest_load(handle, buf, size);
 			break;
+		case TRACECMD_OPTION_PID_SYMBOLS:
+			if (buf[size-1] == '\0')
+				trace_pid_symbols_load(handle, buf);
+			break;
 		default:
 			warning("unknown option %d", option);
 			break;
@@ -3352,6 +3494,8 @@ void tracecmd_close(struct tracecmd_input *handle)
 
 	trace_pid_map_free(handle->pid_maps);
 	handle->pid_maps = NULL;
+	trace_pid_symbols_free(handle->pid_symbols);
+	handle->pid_symbols = NULL;
 
 	trace_tsync_offset_free(&handle->host);
 	trace_guests_free(handle);
diff --git a/tracecmd/trace-dump.c b/tracecmd/trace-dump.c
index 5642f12a..5c6992c8 100644
--- a/tracecmd/trace-dump.c
+++ b/tracecmd/trace-dump.c
@@ -530,6 +530,9 @@ static void dump_options(int fd)
 		case TRACECMD_OPTION_GUEST:
 			dump_option_guest(fd, size);
 			break;
+		case TRACECMD_OPTION_PID_SYMBOLS:
+			dump_option_string(fd, size, "PIDSYMBOLS");
+			break;
 		default:
 			do_print(OPTIONS, " %d %d\t[Unknown option, size - skipping]\n",
 				 option, size);
-- 
2.26.2

